Two-Dimensional Transformation with Minimum Buffering

1. An apparatus comprising: a first circuit configured to (i) generate a sequence of intermediate matrices by multiplying a sequence of original matrices by a predetermined matrix and (ii) write said intermediate matrices into a memory in a first order that alternates between a row order while processing a first of said original matrices and a column order while processing a second of said original matrices; and a second circuit configured to read a plurality of elements of said intermediate matrices from said memory in a second order that alternates between said column order while accessing said intermediate matrices generated from said first original matrix and said row order while accessing said intermediate matrices generated from said second original matrix.

2. The apparatus according to claim 1 , wherein (i) one of said intermediate matrices is generated in said first circuit in each of a plurality of cycles and (ii) each of said intermediate matrices is stored in said memory as either a row matrix or a column matrix.

3. The apparatus according to claim 1 , wherein said memory is configured to both (i) read a row or a column of said elements and (ii) write one of said intermediate matrices in each of a plurality of cycles.

4. The apparatus according to claim 1 , wherein at least two of said intermediate matrices are written into said memory in consecutive ones of a plurality of cycles.

5. The apparatus according to claim 1 , further comprising a third circuit configured to latch each of said intermediate matrices prior to said write into said memory.

6. The apparatus according to claim 1 , wherein said second circuit is further configured to generate a transform matrix by multiplying said elements read from said memory by another predetermined matrix.

7. The apparatus according to claim 6 , wherein said predetermined matrix and said another predetermined matrix transform data in said original matrices either (i) from a spatial domain to a frequency domain or (ii) from said frequency domain to said spatial domain.

8. A method for transform with minimum buffering, comprising the steps of: generating a sequence of intermediate matrices by multiplying a sequence of original matrices by a predetermined matrix in a circuit; writing said intermediate matrices into a memory in a first order that alternates between a row order while processing a first of said original matrices and a column order while processing a second of said original matrices; and reading a plurality of elements of said intermediate matrices from said memory in a second order that alternates between said column order while accessing said intermediate matrices generated from said first original matrix and said row order while accessing said intermediate matrices generated from said second original matrix.

9. The method according to claim 8 , wherein (i) one of said intermediate matrices is generated in said circuit in each of a plurality of cycles and (ii) each of said intermediate matrices is stored in said memory as either a row matrix or a column matrix.

10. The method according to claim 8 , wherein said memory is configured to both (i) read a row or a column of said elements and (ii) write one of said intermediate matrices in each of a plurality of cycles.

11. The method according to claim 8 , wherein at least two of said intermediate matrices are written into said memory in consecutive ones of a plurality of cycles.

12. The method according to claim 8 , further comprising the step of: latching each of said intermediate matrices prior to writing into said memory.

13. The method according to claim 8 , further comprising the step of: generating a transform matrix by multiplying said elements read from said memory by another predetermined matrix.

14. The method according to claim 13 , wherein said predetermined matrix and said another predetermined matrix transform data in said original matrices either (i) from a spatial domain to a frequency domain or (ii) from said frequency domain to said spatial domain.

15. An apparatus comprising: a first circuit configured to write a sequence of intermediate matrices created from a sequence of original matrices into a memory in a first order that alternates between a row order while processing a first of said original matrices and a column order while processing a second of said original matrices; and a second circuit configured to (i) read a plurality of elements of said intermediate matrices from said memory in a second order that alternates between said column order while accessing said intermediate matrices created from said first original matrix and said row order while accessing said intermediate matrices created from said second original matrix and (ii) generate a transform matrix by multiplying said elements read from said memory by a predetermined matrix.

16. The apparatus according to claim 15 , wherein (i) one of said intermediate matrices is generated in said first circuit in each of a plurality of cycles and (ii) each of said intermediate matrices is stored in said memory as either a row matrix or a column matrix.

17. The apparatus according to claim 15 , wherein said memory is configured to both (i) read a row or a column of said elements and (ii) write one of said intermediate matrices in each of a plurality of cycles.

18. The apparatus according to claim 15 , wherein at least two of said intermediate matrices are written into said memory in consecutive ones of a plurality of cycles.

19. The apparatus according to claim 15 , wherein said first circuit is further configured to generate said intermediate matrices based on said original matrices and another predetermined matrix.

20. The apparatus according to claim 19 , wherein said predetermined matrix and said another predetermined matrix transform data in said original matrices either (i) from a spatial domain to a frequency domain or (ii) from said frequency domain to said spatial domain.